The Markus way with dichotomies: Corrective and distributive justice

Another Aristotelian dichotomy – Distributive justice and corrective justice

For the rest of my time I will speak of another dichotomy, which Aristotle discusses in the Nicomachean Ethics (Aristotle, 2011). I ask: what can George’s approach to poiesis and praxis tell us about the categorical distinction Aristotle makes in Book 5 between two distinct types of justice: distributive justice and corrective justice? ⁴ By George’s approach I mean an approach that relativises the sharpness and usefulness of any dichotomy by always understanding it within a particular historical and cultural framework. And if the dichotomy has any staying power this is not due to its philosophical truth but to the social context which provides the material for its application; namely, culturally specific expectations, social descriptions of action and specific standards of evaluation.

In Aristotle these two types of justice are distinguished first by their scope: distributive justice is exemplified by state action allocating benefits and burdens to agents in general (distributing public funds is Aristotle’s example). Corrective justice is the agent-specific application of justice operating in the interactions or dealings of one person with another (either voluntary transactions, contract, or involuntary, tort or crime). Second, the types of justice are distinguished by the applicable criteria. In distributive justice it is proportionate to the criteria of merit and this will vary from polity to polity. In a democracy it will be the criterion of equality, says Aristotle; in an oligarchy, wealth; in an aristocracy, excellence. In corrective justice there is only one criterion and it is arithmetic. One person has affected in some way what already belongs to another. The law looks only at the difference created by the damage done.

For it makes no difference whether a decent person has defrauded a worthless one or a worthless person a decent one, or whether the adultery was committed by someone worthless or decent; the law pays attention solely to the difference created by the damage done. (Aristotle 2011: 163)

I don’t pursue any of this but ask about the orienting role of this dichotomy in two contexts – Kant’s discussion of private right as a substantive account of corrective justice and the relationship of corrective justice and distributive justice in Rawls’ discussion of political justice. As will become clearer, I am focusing on that aspect of the dichotomy that treats the criteria of the two types of justice as distinct.

Corrective justice and Kantian right

There is an influential school in private law theory which understands Aristotle’s analysis of corrective justice along the lines of Kant’s principle of right. ⁷ It is unclear why this school came to prominence in North America in the 1980s. But any explanation would point to its linkage with liberal conceptions of justice and its aim to offer an account, originally of tort law, which rivals instrumentalist accounts of law (law and economics, public policy, welfare justifications for law). ⁸ The point of tort law, it is said, is to respond to interferences with freedom, i.e. private wrongs; not to compensate plaintiffs (via efficient loss distribution) or deter defendants. Whatever relevance these factors have must be seen through the bipolar structure of free agency and the protection of rights. ⁹

If you recall, Kant divides the law part of the Metaphysics of Morals (Kant, 1991) following Roman law categories into private right, which he links with corrective justice, and public right, which he refers to as distributive justice. ¹⁰ The organising principle of corrective justice, the principle of right, is the right of independence, the right to set your own choices unimpaired by other wills; unimpaired by the need to take account of others’ needs or ambitions. We should be free to do whatever we want, as long as we don’t wrong others (i.e. interfere with their freedom to do as they want). Free actions may harm others, alter the context in which they act (this is inevitable) but the only justification for corrective justice and coercive law is when we wrong others, i.e. ‘wrongly’ subject others to our will.

The argument is that contract law, tort law and the law of restitution for unjust enrichment must be understood as based on this distinction between wrong and harm. Wrong actions are violence, taking, threats or deception where you draw in another to serve your purposes. For these actions violate the embodiments of another’s free will, namely their bodily integrity, their acquired entitlements (as rights to property or contract rights) or their free agency. When a wrong occurs, corrective justice demands that things be equalised and the gain is taken away. Sometimes this can be seen as a factual equalisation – you have taken my possession or I have mistakenly paid you money; here the law makes you give these things back. But in many areas of corrective justice the defendant has not benefited factually, say in negligence law; here the gain is seen as normative. The defendant has a normative gain as he has breached the duty not to offend the plaintiff’s rights.

Kant saw his analysis as a metaphysical inquiry into free agency; an inquiry that cannot be based on anthropology but can still be applied to it. The inquiry is into the grounds of obligation, not the circumstances of its application, which is not independent of experience (Kant, 1991: 44). ¹¹ The principle of right (which separates lawful from unlawful determinate activity) is itself indeterminate. For general rules cannot specify the particulars that fall within them. The contemporary Kantians agree with this ¹² but claim that the distinction of interfering with another’s means (wrong) and merely changing the context (harm) provides the necessary conceptual resources for any proper application.

At one level this is a reasonable description of private law – what we need to assume to make sense of the practice. The rules are best understood it is said intrinsically, as a concretisation of the wrong/harm distinction rather than extrinsically as based on its function, say providing welfare or deterring wrong-doing or efficient loss allocation. Achieving corrective justice is valuable in itself. This agent-centred approach emphasises the bilateral nature of transactions. You recover from the breaching party, not someone else; the relevant considerations are related to the parties. ¹³ And corrective justice reinstates the status quo. You are not here creating a just distribution anew but reinstating the existing allocations. Corrective justice both presupposes the existence of entitlements ¹⁴ and limits the relevance of questions of justice not connected to the wrong.

Let me make this point via the anecdote of Cyrus and the coats from Xenophon’s Cyropaedia (Xenophon, 2001). David Hume introduces this example in Appendix III to his Enquiry Concerning the Principles of Morals. Hume says (1983: 94):

Cyrus, young and unexperienced [as Hume puts it], considered only the individual case before him, and reflected on a limited fitness and convenience, when he assigned the long coat to the tall boy, and the short coat to the other of smaller size. His governor instructed him better, while he pointed out more enlarged views and consequences, and informed his pupil of the general, inflexible rules, necessary to support general peace and order in society.

However, what is made clearer in the original is that there had been a prior distribution and the big boy took the coat of the small boy and Cyrus thought it better to leave things as they now were, i.e. this was his response to a breach of corrective justice. ¹⁵ The point of the example for us is perhaps to pick out one way in which corrective justice and distributive justice are conceptually distinct; namely, once we have found that the plaintiff has been wronged we should not tinker with this to bring about distributive justice. ¹⁶ But in working out if an agent has been wronged, the concerns of distributive justice, as well as other concerns, may well be relevant.

This last point is denied by the Kantians. For they argue that all relevant standards of action must be justified by rightness reasons grounded in interference with individual freedom (i.e. in the same type of corrective justice reasons). ¹⁷ But can the wrongs of private law all be explained as interferences with individual freedom? What of other rightness norms, other components of interpersonal morality, associated with inhumanity or honesty or exploitation? And can the wrongs of private law be understood in their creation or at the point of application without some consideration of what is useful or what is the law’s purposes? Let me address these points with brief examples drawn from tort, contract and unjust enrichment.

Duty of easy rescue ¹⁸

In the Kantian understanding, right is presented as a system of mutual non-interference. Here the distinction between misfeasance (action) and nonfeasance (inaction) is seen as a central organising principle. ¹⁹

A testing example (in theory if less so in practice) is the failure to rescue or aid another in circumstances in which you have not contributed or are not in a special relationship, such as parent/child. ²⁰ This failure, it is argued, cannot be a tort (even if the recue is easy and without risk) as the bystander has not interfered with another. To impose liability would be to confer a gratuitous benefit on another. For in pursuing your ends, you can be indifferent to the needs or welfare of others. ²¹

The Kantians would allow criminal law to impose a duty of easy rescue and many states, particularly in Europe, have done this. Here the Kantians argue the duty is owed to society at large, like the duty to pay taxes or duties of conscription. In other words, the problem should be seen as an aspect of distributive justice, not corrective justice. Any breach is owed to society as a whole and not to the victim.

But why cannot private law be used to promote values other than free agency? Why cannot tort law’s understanding of wrong be used as a way of imposing proper standards of interaction, in this case the promotion of welfare in an emergency through discouraging callousness and encouraging samaritanism. ²² Why shouldn’t these reasons count for something in this context?

Contract

A focus on the bipolar relationship presents contract law as controlled by the will of the parties. But understanding their agreed will invokes matters assumed by or external to their agreement. For example, the parties through their actions cannot control either the legally established criteria that are needed to make a binding agreement or the social understanding of matters that may render the agreement defective – duress, mistake, misrepresentation, undue influence, etc.

It could be said that these types of external matters can be contained within the idea of a bipolar agreement. They set out what is needed to make a legal agreement and point to matters that might defeat a meeting of minds. But there are other ideas based on other notions of fairness that may make us question the enforcement of the agreement. For instance, has one side taken advantage of superior bargaining power, has a standard form contract deleted consumers’ rights or imposed predatory or unconscionable terms, are the terms discriminatory? If these are reasonable descriptions of what is going on these matters could be relevant in the assessment of any particular exchange. ²³ And none of these matters could be called a direct interference with free agency.

The recipient of a mistaken payment of money must pay it back

A bank has mistakenly paid money into your account. A shopkeeper has mistakenly given you too much change. The law demands that the money be returned. This is the core example of unjust enrichment. But any account of this which focuses on the payee’s agency struggles with explaining this rule. ²⁴ For at the time the money is mistakenly paid the payee has done nothing wrong. When the mistake is discovered perhaps you can say that the refusal to return the money is a wrong, or the payee has no loss at this point of time if it is returned. But both of these claims assume at this second point of time that the payee is not entitled to the money – the very thing that has to be established.

Perhaps what has gone on could be described as the payee has accepted a benefit in the context where you can assume (without further facts) the payor did not intend to make a gift of the money. But what does acceptance mean here, as the payee is initially ignorant of the mistake? It is not like knowing someone is building on your land and keeping quiet and accepting the benefit.

Here we have a rule in most legal systems, whose justification cannot be given from within Kantian corrective justice, as the payee has done no wrong. ²⁵ The unjust enrichment wasn’t wrongfully created but nonetheless it would be wrongful now not to reverse it. ²⁶

What we learn from these examples is that the bilateral structure understood as Kantian corrective justice cannot on its own determine effectively what is normatively relevant for the content and application of private law? As Hegel would put it (and did put it), it cannot determine on its own its juridical relationships. ²⁷ If we want to have a way of describing what action was performed, and see what counts here for us, we need to refer to the social structure of justification in which the alleged wrong took place. A structure that will contain more ideas about justice than an answer to the question – has one agent interfered with the will of another agent? The Kantian way of cabining corrective justice is misleading and unhelpful.

Corrective justice and Rawls

From the perspective of distributive justice, what is the relationship between corrective justice and distributive justice? Or more specifically, what is the place of private law in Rawls’ account of political liberalism?

Distributive justice for Rawls is focused on providing a fair social framework by regulating the inequalities in life prospects that arise from social starting positions, natural advantages and historical contingencies (Rawls, 2005: 271). His well-known principles of justice specify the fair terms of cooperation for a free and equal people. These principles are not directed at particular transactions but at the background social framework, the basic structure, in which individual activities take place. That is why In A Theory of Justice (1971) and Political Liberalism (2005) Rawls has nothing directly to say about corrective justice and little to say about other criteria of justice. ²⁸ But what are the implications of his account for corrective justice? How might the principle of equal basic liberty, the principle of fair equality of opportunity and the difference principle apply to private law? Recall that these principles are said to operate not just at the stage of the original position or constitution making but at the subsequent legislative and judicial stages of law making and law application (Rawls, 1971: 195–201).

One answer to this might look at how Rawls himself understood the basic social framework. In Justice As Fairness (2001) this is said to include the political constitution and the central features of its economic and social systems. ²⁹ This is further explained in the essay ‘The Basic Structure as Subject’, where Rawls sets up a division of labour between two kinds of social rules and the different institutional forms in which they are realised. ³⁰ On the one hand, there are the rules that apply to the basic structure, and here Rawls includes the operations of income and inheritance taxation designed to even out the inequalities in property ownership. On the other, there are different rules that govern transactions and agreements between individuals. Here Rawls gives the example of the law of contract (Rawls, 2005: 268). He goes on to say (Rawls, 2005: 283):

The difference principle holds for example for income and property taxation, for fiscal and economic policy. It applies to the announced system of public law and statutes and not to particular transactions or distributions nor to the decisions of individuals or associations, but rather to the institutional background against which these decisions and transactions take place.

But if private law as part of the state’s legal system is part of the basic structure, how should Rawls’ principles of justice guide the development of this area of law? Or to put it differently, what scope here for other values than those of distributive justice, i.e. non-distributive principles of justice? How can we link the values of distributive justice to the values associated with private law and corrective justice – free and reliable exchange, rightful ownership, personal security and the ability to respond to those who have wronged us. ³³

This is not a topic that has received much attention in Rawls’ commentary. Here are three responses. Arthur Ripstein sees it in terms of public and private law. While contract law and tort law are possibly part of the basic social framework, as private law they should be seen within Rawls’ framework as having ‘a certain kind of independence’. Particular transactions should be judged by the remedial norms of individual responsibility and not subordinated to distributive justice unless the aggregated effects of these transactions lead to distributive injustice – by worsening the position of the worst-off group, for example. ³⁴ ‘But generally, society should only use such means that are consistent with the freedom of separate persons (Ripstein, 2006: 1403). Despite the escape clause of ‘a certain kind of independence’ this approach maintains the dichotomy of distributive justice and corrective justice, now understood as the classification of legal ideas and criteria into those of public law and those of private law.

For Peter Benson the way to see it is that corrective justice has lexical priority over distributive justice. By this he has in mind that, as in Kant and Hegel, corrective justice has conceptual priority. You must understand what is involved in free agency in the abstract before you can understand how it might be realised or what follows from our embeddedness in social relationships in, say, a fair system of cooperation. For Benson all that follows from this is that provided the legally enforced claims of private law have the form of corrective justice (respect, say, our equal capacity for rights (of ownership), our innate right to bodily integrity), then the content of these norms can come from distributive justice, or other relevant principles of justice, based in legitimate needs, say. Lexical priority does not mean that corrective justice has principles that are normatively antecedent to or limit in principle the use of criteria based on notions of distributive justice. ³⁵

Samuel Freeman in his recent Liberalism and Distributive Justice (2018: ch. 5) has the most extensive discussion of this topic. He argues that Rawls’ view, properly understood, is that property law, contract law which affects market transfers and tort law are all part of the basic social framework. Property rules and contract rules should be devised that do not interfere with or burden and even promote the effective exercise of basic liberties or fair opportunities ‘to an adequate degree’. In other words, these rules should enable persons to exercise their basic liberties and, at least with economic contracts, the rules should be consistent with fair equality of opportunity. ³⁶ The difference principle applies to the specification of property rights and the rules of economic contracts. ³⁷

Tort law is subject to the principles of justice, in as much as tortious wrongs unremedied can undermine the effective exercise of rights and opportunities. But the difference principle, argues Freeman, does not apply to tort law. ³⁸ Compensation should be determined by plaintiffs being returned to the prior position, not by some recalculation that reflects a fair distribution of the social product. And in as much as aspects of private law turn on issues of personal responsibility, how can these be governed by principles of justice that focus only on distribution? ³⁹

What we see in these three accounts is an acknowledgement that distributive justice might well be relevant to questions of corrective justice; and, in the case of Ripstein and Freeman, some attempt to structure in general terms how this relevance may come into play.

Conclusion

What of the orienting role of Aristotle’s distinction between the two types of justice? If you recall, there were two claims. First that there is a difference in scope; and here there remains for us a clear enough distinction – the general allocation of benefits or burdens as against the local application of justice after the wrongful dealing of one person with another. But it is the second and more interesting claim that we have been exploring – that in the applicable criteria, there is a categorical difference between distributive justice and corrective justice.

We should approach this categorical claim in the Markus way, for the dichotomy does not itself pick out distinct criteria and distinct concerns. For example, the criteria applicable in corrective justice are not helpfully thought of by us as arithmetic (as addition and subtraction or as re-establishing the mean) as Aristotle has it, and nor are they well-defined by Kantian right, as an interference with free agency. This is a too limited and dogmatic understanding of the dichotomy.

The proper way, the Markus way, for seeing an application of corrective justice is to look through its form and see that it takes place within a particular social context, a living ethos, which makes available cultural expectations and evaluative criteria – ways of describing what has gone on, often contested, and legal and social standards for evaluation of this conduct. ⁴⁰ Which actions can be imputed to the defendant? Here distinctions must be made between harms which are intentional, reckless, negligent and accidental. And thought must be given to what is significant about these actions and from what perspective? For in assessing these actions should we follow standards of rightness or standards of efficiency (promoting law’s purposes) and how may these demands be reconciled and in what way? All these questions are context sensitive and answers cannot be teased out of the idea of corrective justice alone, but only with reference to the social understandings and the structure of justification in which the action took place. We have to see what is relevant and weigh these matters – where what is relevant and how things are weighed depends on matters local to the socially embedded agents and the situation.

In short, an application of corrective justice can only be made on the basis of practical reason with reference to the concrete situation. Understanding the law, establishing the wrong and what should follow means thinking of the case along with the law so that what is concretised is just, as far as this is possible. And in doing this we cannot systematically keep the ideas of distributive justice (fairness, need, desert, welfare, usefulness, etc.) distinct from the ideas of corrective justice.